Possible eruption at Bárðarbunga

The almost obscenely inconspicious volcano of Bárdarbunga seen in all it's icy glory.

The almost obscenely inconspicious volcano of Bárdarbunga seen in all it’s icy glory.

First of all let me say this, it is only IMO and Almannavarnadeild ríkislögreglustjóra that has the authority to officially declare an eruption in Iceland and issue warnings. Untill they have officially released a warning for an eruption there has been no confirmation.

That being said, below follows what I think is happening currently at Bárðarbunga.

Official warnings

There has so far just been two warnings issued and that is for a potential Jökulhlaup and about a magmatic intrusion. If a medium to large eruption takes place a large Jökulhlaup could burst out and inundate land around the rivers and could be a potential danger to the locals. Here are the warnings issued by Almannavarnadeild ríkislögreglustjóra:

“The National Commissioner of Police and the District Commissioner of Police at Hvolsvöllur and Húsavík have declared a Civil Protection Uncertainty phase due to unrest in Bárðarbunga.”

“Scientists says August 17 at 1000. The earthquakes that began at 3 am yesterday morning continue. The activity has shifted and is now in two clusters, north and east of Bárðarbunga. Scientists believe that the activity is caused by magma activity in the Earth’s crust. No signs yet that the eruption has started. Nothing can be ruled out that the chain of events that is now underway will lead to an eruption outside the glacier or beneath it. Eruptions under glaciers could lead to flooding in the rivers that flow from it. Scientists and the Civil Authorities closely monitor the situation and the uncertainty warning is still valid.”


Earthquake distribution map from IMO.

Earthquake distribution map from IMO.

The earthquake swarm continues unabated in and around the volcano. It has been noted that the activity is centering on several place and the picture is confusing to put it mildly. So far there have been 853 earthquakes in total with 108 earthquakes between M2 and M3 and 7 earthquakes larger than M3.

The 7 M3+ earthquakes have taken place at different spots, 4 of them is along the Dyngjuháls fissure swarm near Kistufell. 2 are east and east-southeast of Bárðarbunga, and one is a Kverkfjöll. The first four at Dyngjuháls points towards a possible eruption in this area, it is a common spot for Bárðarbunga fissure eruptions.

There are though 3 more earthquake foci, one running ESE from Bárðarbunga, another running towards Kverkfjöll and the third is running from Hamarinn towards Grimsvötn.

Harmonic tremor

Image by IMO, the current seimic and tremor signals at the DYN SIL-station.

Image by IMO, the current seimic and tremor signals at the DYN SIL-station.

So far there is no clear harmonic tremor signal on the SIL-network; there have though been what looks like pulses that could have been gas release in rising magma.


Hamarinn GPS (HAFS) from Sigrún Hreinsdottirs page. Note the sudden jolt southwards, westwards and up. Click on image for large scale image.

Hamarinn GPS (HAFS) from Sigrún Hreinsdottirs page. Note the sudden jolt southwards, westwards and up. Click on image for large scale image.

Image from Sigrún Hreinsdottirs page. Notice the rapid jolt northwards. Click on image for large scale image.

Image from Sigrún Hreinsdottirs page. Notice the rapid jolt northwards. Click on image for large scale image.

I normally say that you need at least 3 data points to be sure of a GPS-trend, and that 5 is preferable. In this case we have 1 data point at two different GPS-stations. Even though it is too little the amount of motion and the direction of motion is of such a dignity that I think it is correct (sanity check).

What they are showing is a magmatic intrusion at depth in or around the central volcano of Bárðarbunga. Dyngjuháls GPS is going in Northerly direction and Hamarinn is being pushed to the south as you can see on the two images.

One thing here is that the GPS points to rapid inflation, and that would mean that as those data points where collected no eruption had started since that would most likely give a more stable signal, or even rapid deflation.

Eye witness reports

There are currently no eyewitness reports or no reports of anything having been spotted during the flyovers.


At this point I am fairly convinced that an eruption either has started or is about to start. The signs are just a bit too overwhelming right now.

I think that what we will be seeing is an eruption along one of the radial fissures or upwards through the Dyngjuháls main fissure swarm of Bárðarbunga. There is also a chance that we will see Dyngjuháls and one of the radial fissures erupt simultaneously and there is a possibility that the central volcano also will erupt.

It is rather unlikely, but not completely impossible that the current activity will lead to a larger rifting eruption up in the northern part of the Bárðarbunga fissure swarm (Dyngjuháls). I repeat that this is currently unlikely.

Here is a link to our Bárðarbunga links page where you will find all sorts of goodies about the volcano. The author behind the Volcanocafé Bárðarbunga is Spica. This page will be continuously updated so look there for new material daily.



573 thoughts on “Possible eruption at Bárðarbunga

  1. From Jon on FB: According to Mbl.is it is now believed that magma is on the way up to the surface at two locations in Bárðarbunga volcano

  2. I am more interested in the possible effects of an ash eruption on aviation. I see today the jetstream is slap bang over Iceland with a flow towards UK and Europe. Here in the UK it is the summer bank holiday weekend begins Friday – the busiest weekend for aviation in the year. I can see the perfect storm of ERUPTION + JETSTEAM = CHAOS! Any thoughts on how high this volcano erupts, type of material it could eject etc..?

    • If its Kistufell, then your guess is as good as mine.
      It is a posibillity of an explosive eruption at Kistufell, but all the rest is just speculation.

    • The forecast for high altitude winds seems to be the same for as far as it reaches. Any (still hypothetical) ash plume would be carried south or southeast. Straight into the busy heart of European airspace.

    • If it is a fissure eruption it is less likely to be ashy compared to an eruption out of a central volcano like Bárdarbunga itself or Kistufell.
      I am reticent to give a definite answer to what will happen.
      I defer the question to the official warnings comming from London VAAC and IMO on this.

      On a general note, an eruption would most likely affect air traffic, but that is how far I would ever go commenting this. Over to London VAAC it is.

    • In the Info about Badabunga, up in the menu ( https://volcanocafe.wordpress.com/info-on-bardarbunga/), there are Jetstream pages and earth wind map ( animated). Winds would not be really good too, in case something like Eyja should happen when the ash did not reach the stratosphere because it only went up to 9 km. Winds are blowing down over England all the way to the Canaries.
      As Carl said. it might take quite some time till the eruption breaks through the tick glacier, or it could stop before this happens or … we ll see with time.
      Material: other will know better than me, but volcanic material always has quite some silica in it. What Eyjas ash looked like can be seen in https://volcanocafe.wordpress.com/eyjafjallajokull/

      • The question is will Bardabunga Magma be like Eyjafjallajokull or more like Grimsvotn which didn’t carry as far. I am leaning towards ash that is more like Grimsvotn but who knows

    • Let me state one more thing as a layman. regardlessly which composition the ash actually has, it is never safe for a plane to fly through a cloud of ash. the turbines are hot enough for the material to melt and harden as soon as the engines are turned off.

  3. Hello.

    Já, Já. I am now Upset. Not in a good moooð.

    Who is nocking me!?
    I ate tourists.

    Grinning and hissing.
    Grinning and hissing.
    Grinning and hissing.

    Swollen sick kicking like a sore owl.

    Not in a good moooð. moood, moood

    But I love you guys.

    • That thing is really quite big. Hiked there about two weeks ago, but the weather was so bad that i couldn’t take any pictures.

      there are a lot of lavafields in this area, especially to the west of Kistufell, also some Pillow Lava can be found there, that normaly form underwater or under thick ice.

  4. Bardabunga Earthquake Animation 16-18/08/2014
    the data comes from the “listi” pages

    Don’t hesitate to play with the definition (up to 1440) or go on YT to see it directly.

    There are 2 event by event sequences, with views from the south then from the East. Then there is an horizontal rotation with all the quakes.

    Finally there is a vertical rotation and return.
    The color of the quakes gives the date as shown on the left of the colorbar. Terrain elevation is given on the right of the colorbar. On the first sequence, the magnitude if the earthquakes is given by the scale on the right of the plot.
    Data from IMO and NOAA, made on Gnu Octave.

  5. I was just looking at the tremor graphs from IMO. Is there any reason why Melhnauser (mel) jumped up on 14th Aug? It then jumped with all the others on 16 Aug. I’ve no idea where it is on the map – but I think it’s interesting.

  6. http://in.reuters.com/article/2014/08/18/iceland-volcano-idINL5N0QO2NV20140818

    “Met Office seismologist Martin Hensch said the risk of any disruptive ash cloud similar to the one in 2010 would depend on how high any ash would be thrown, how much there would be and how fine-grained it would be.
    Eurocontrol, the Brussels-based agency responsible for co-ordinating European airspace, said in a statement it was aware the Icelandic Met Office had revised the status of the volcano and it was following the situation closely.”

  7. You have to remember that the quakes are only held for 48 hour. So now the 48hrs have passed, we are loosing the beginning of the swarm, yet gaining more. = profit

    • Is the man for real?
      Eyjafjallajökull was in 2010 so he should not forgot that it had more than two weeks of earthquakes before erupting.

      • Regardless of what Páll thinks it seems that Bárdarbunga has a different opinion. So did by the way Fimmvörduhals, Eyjafjallajökull, Katla, Grimsvötn and now Bárdarbunga. Telling volcanoes what they should do has not really worked out for Páll.

      • I guess he mean, that it is the longest time, between onset of the tremor as an strong indicator of magma movement and definate signs of an eruption, such as higher river outflows, higher gas measurements or even a widnessed eruption.

      • I’m not an expert, I’m a scientist but not in this area, not even close. But I think that in the grand scheme of things, out of all the millions of years of earthquakes on earth (and Iceland) we’ve only collected data for a tiny fraction of them, so we can’t really draw strong conclusions from what has happened in the past 30 years.

    • Perhaps the caveat here, and what he is trying to say, is that it’s unlikely to have such an intense swarm without an eruption? I can’t remember the daily quakes within those two weeks that led to Eyja to compare intensity but we did and do see 100s and 1000s of quakes at other sites without an eruption, haven’t we recently in Ecuador and Also at Bob?

      I can easily believe this may turn into just one of a few intrusions before the big event but the statement about it being an overly lengthy run-up seems contradictory given the known examples unless he means intensity and HT. Lastly, we all know volcanoes are fical creatures and don’t always behave as expected so even if the statement is generally true, it doesn’t mean it will be this time round!

      • yep. quite right. I remember the run-up to Fimmvorduhalsi being just as intense but you’d need to measure the cumulative seismic energy to put that into any meaningful context. This one is possibly larger already. At least in terms of the number of quakes and repeated swarms, Eyaf was much longer than this. And yes, most swarms actually don’t lead to an eruption. But this one has that whiff to it. It’s evolving and it’s not waning. I am betting on an eruption.

        There I jinxed it. Sorry everyone. Blame me if it doesn’t happen.

  8. Almannavarnir @almannavarnir · 2 h

    Today Icelandic Civil Protection is meeting with Embassies officials and key domestic stakeholders due to seismic activity in #Bardarbunga

  9. I don’t think that Bardabunga, Kistufell or even a small rift in north-western vatnajökull is erupting at the moment. There are strong signs, which point to an eruption, but there are also good signs that this is only an intense dike intrusion.

    If one look at the different signs, that we see, than we see a very intense earthquake swarm, which migrates from the caldera and splits into to areas, one along the Dyngjuháls fissure swarm and one east of the caldera. We see a north/south and east/west motion at different gps-stations, but no real up- or downlift motion. We don’t see higher river outflows from the area and we don’t see higher or typical volcanic substances and gases in the rivers. [A couple of comments before, there was the question, why iceland has no gas measurements. The simple answer is, that it is much easier and much better to get good and reliable values from gas measurements in rivers than from the air. For air measurements you need higher values, the measurement station has to be close to the point, where the gases are came out and of course strong winds causes a fast distribution of the volcanic gases. Besides this, most of the iclandic volcanoes are beneath ice. Therefore it is the easiest way to measure the gases in the rivers.]

    What point to an ongoing eruption at the moment?

    – Intense earthquake swarm and the point that other volcanoes in iceland don’t need much earthquakes during the begining of an eruption
    – The distribution of the quakes. During the first four weeks of the El Hierro eruption, we had very strong earthquakes, which were the result of the discharging magma-chamber.
    – Strong gps-motion to the north/south and east/west without an uplift component
    – tremor

    What point against an eruption at the moment?

    – Not all intense earthquake swarm lead to an eruption. Compare it to El Hierro, where we had several very strong dike intrusions after the eruption and until now not a second eruption (I know, that it is very hard to compare it to El Hierro, but I know, that al lot of the readers are familiar which el Hierro).
    – Earthquake distribution looks more like a dike migration
    – gps-motion, don’t show a downlift
    – no higher river outflows or higher gas values in the rivers

    I would love to see an eruption at Bardabunga, but until now I think this event is only a very intense dike intrusion, which could probably lead to an eruption in the near future (couple of days or weeks) or only in the next years to decades.

    • I think the idea that Iceland volcanoes don’t rumble for very long before an eruption is based on a very limited data set. Sure, Grimsvotn and Hekla have gone right from nothing to eruptions in a surprisingly quick fashion.

      But historically, we’ve known that volcanoes like Katla act more normal in that there is a long period of swarms prior to an eruption. We have reason to believe the same was true for Laki, and the Krafla fires had intense swarms prior to erupting as well.

      This is mostly just personal speculation, but I’m not really buying that we’re going to see an eruption here within 24 hours. I think it’ll erupt eventually, but I think it’s going to rumble and rock for a while before we see any eruption. I have a feeling we’ll see the area expand and propogate outwards bit too if this is going to be a fissure eruption of any sorts.

      • I think the field is wide open (excuse the metaphor).. I agree with what you say about the limited dataset.

        Hekla and Grimsvötn are both pretty open systems so they pop their top with little warning, kind of like Etna, sometimes with even less warning than Etna.

  10. Hey. I’m a person who loves volcanoes, so i’m following this earthquake activity in Iceland pretty closely now.

    Now, as there is a web cam (that updates every 10 mins) here: http://vedur2.mogt.is/grimsfjall/webcam/ , i do see that there are some black spots on the ice on the left side of the pictures there. Is that ashes from other volcanoes that have been in the same area earlier or what?

  11. If IMO is only showing the last 48 hours of the quakes, is there someplace where we can see the total number of quakes since this swarm began?

  12. Phew. Windy 4.2 hrs flight. No eruption yet. Coastguard Helo was there and some other planes. Nothing to be seen yet. There were clouds on the north-east slope, and we did not fly directily over. No point. We contained us by looking better at west side of glacier and in the Laki – veiðivötn areas. Very impressive large blast craters. Some a mile long! Some are filled with water. Did also photo time over Landmannalaugar.
    Maybe activity will suddenly shift and that areas are gone in an instant (and different landscape taking over). First time I have taken detailed look at individuel craters in Laki line.
    Very impressive.
    So flyover?
    No. Rather a flyby today!
    Flyover at better time.


  13. Interesting development!

    What kind of ash-cloud, aviation-wise, is an eruption likely to produce?

    I work as an Area Centre Air Traffic Controller in South-Western Norway (western part of airspace forms a line from approx. 0E/58N-> 0E/63N), and during the Eyafjellajøkul eruption our airspace went haywire with tons of redirected flights….

    So whats the likely consequences this time around? 🙂

    • I will make an official statement in a little while. At least official from VC.
      This will cover the air issue and be a general update.

    • Someone else will help you better with that, since I am a total newbie, but I would say that if at some point Kistufell will erupt, it is hard to say because nobody really knows what magma exactly is under the volcano or how its sill system actually looks like. If it has it own magma chamber, than it should be packed with some explosive stuff, as Carl said earlier, due to to possibility of evolved magma being present. I would say that since Kistufell is at a central point, it probably had occasional small intrusions, keeping the magma there at some steady temperature, but not enough to cause pressurization of the chamber, shall there be one. Just my newbish speculation. 😉

    • To be honest at the moment we do not know which type of ash could potencially affect the European airspace. We do not know how large the eruption might be, or how long. Let us just say that so far a large eruption is possible but at the moment there are no signs pointing to this specifically. There is a strong possibility that a future eruption would most likely be largely efusive (not so much ash), especially as the swarms are learning towards a ice-free region.

      The size of grain is of course the important factor in determining how far away could a ash cloud blow. The ash was unusually fine grain in the case of Eyjafjallajokull, because usually volcanic ash tends to be coarser grain which falls much closer to the site of eruption. This we can’t predict at the moment.

      What will be most important, and like in the case of Eyjafjallajokull in 2010, it is the wind direction. That is critical, as any northern wind would push ash into European airspace. For that, the best is to consult any long term weather forecast mathematical models. They would predict prevaining winds for the next 3 weeks.

  14. I made a gif of the quakes over the last 3 days – it’s interesting to see how this has expanded. It seems quakes above Bardarbunga itself have slowed down. At Kistufell, they’ve intensified, and spread slightly north.

    The most interesting part is that line to the east, where you can see a fissure line expanding and starting to propogate further north.

    quake gif

    I think that eastern fissure will keep expanding further north as the swarm continues.

  15. Important update.

    I can now tell you, at this level of activity, we still need to wait several weeks until we see a proper sized eruption. Unless a string of M5 quakes hits, then an eruption would just be hours away, if that number of M5 quakes is near 10 (or same equivalent cumulative seismic energy), or one M6 event or near that.

    Why is this?

    The GPS stations at both sides of the intrusion taking place at Bardarbunga show a lateral displacement of around 2cm per day (1cm/day in both directions).

    During the last rifting event at Bardarbunga, the rifting displacement was around 1.5 meters (150cm) as the fissure opened, over some 15km. Let’s say we expect an event about half of it. To be on the conservative side. And 7km is just the same as Gjálp.

    At this rate, we will need 35 days, to see a displacement of 75cm as the rift opens.

    Total energy combined is probably equivalent to a M5. Until yesterday I calculated total energy to be about M4.5. Today it nears M5 (total cummulative seismic energy). That equates to a displacement of a few cms, which is what is actually taken place. Which is too little!

    One M6 equates to a displacement of 50cm, so it is near what it is necessary to see such a fissure ripping apart of sufficient lenght; or a string of 10 quakes at M5.

    Thus until this moment, only about 1/10th of the threshold energy has been released.

    So this gives 10 times longer time, until we see an eruption of the type I assumed, if the current level of activity keeps.

    But a few M5 quakes would quickly speed the process, to only a few days necessary.

    So I predict to see either more 9 times of intrusions with the same sort of intensity (if such activity would now stop – which seems to be unlikely at the moment), or this intrusion continues for more 35 days (until mid September), or the energy increases until the level of several M5 or one M6 event (which at current level of increasing activity is just a few days ahead). If this intrusion continues for some more days but then dies down, then we would still need another one or few repeats of same intensity, until a fissure event.

    Consider this the threshold.
    Now, you know when to expect the start of the eruption.

    Simple maths.

    • Good math. 🙂 Perhaps this is the same reason why Mr. Páll has his opinion the way it is. 🙂
      But that would be for a fissure event.
      How about Kistufell? Any possible math on that?

      And P.S.: Just in case, here are the 300mb (~9.200m altitude) geopotential height contours (pattern) and wind forecasts from ECMWF.

      Pretty much the worst possible weather pattern you can imagine for Europe, just in case an explosive eruption actually happens.

    • I agree we have a bit to go before we see an eruption, but I’m not sure we’ll need a full 1.5 meters of rifting before we see an eruption.

      One thing to keep in mind – the eruption itself likely pushes a rift boundary much further apart by itself. Think of it this way – with the 1.5 meter crack from the previous eruption, do you really think that the entire rift needed to separate by 1.5 meters before an eruption started? That’s a huge gap, and given tons of pressure and magma ready to let loose, it would seem silly that a gap that wide would be necessary to start an eruption.

      It’s much more likely that it started erupting earlier, and the eruption process helped spread the rift open further. Remember, many of the rift eruptions here seem to operate on a feedback loop, one which encourages more decompression as more magma is released.

      Also, think of a simple model such as this – if you’re in a submarine, and you open the door – the water rushing in will push the door open all the way simply due to the pressure coming through. Replace the door with a rift opening, and the water with magma and you get similar affect. Now, I know that’s an overly simplistic analogy, but I would have to imagine it applies here as well at least to some degree.

    • From the math point of view correct, but from the volcanic point of view, I think it is not that simple and not correct. Remember Hekla or El Hierro. Hekla had a few km long fissure without intense earthquakes. El Hierro had intense dike intrusions without an eruption.
      One M4 or M5 event at the right place with the right magma pressure could be enough for the beginning of an eruption. In almost the same manner several M5 events could be lead to no eruption at all. The only think what we can say is that more or intense earthquakes increases the likelihood of an eruption.

  16. Quakes are also not updated correctly. Besides the M4.5 event in 2:37, there was two quakes around M3.5 at 10:44 and 15:07. The list is just a mess, these two quakes should be revised.

    • They often are not in correct places. Open list in browser, and let browser search look up this minutes, thay likely are just where they were dunped after recalc.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s